Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-2B
    4. Residual Stream - 16k
    5. 4-GEMMASCOPE-RES-16K
    6. 12251
    Prev
    Next
    INDEX
    Explanations

    terms related to the act of restoring or repairing something

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 4-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_4/width_16k/average_l0_124
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.4.hook_resid_post
    Hook Layer
    4
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     المعيارى
    -0.75
    plotlib
    -0.70
    CHAFT
    -0.69
    chengladbach
    -0.68
     EClass
    -0.63
    endphp
    -0.62
     PHILOS
    -0.62
     kumpulan
    -0.62
    CharStream
    -0.61
    EventLoop
    -0.61
    POSITIVE LOGITS
     restore
    2.47
     restoration
    2.38
     restored
    2.32
     restoring
    2.23
     Restore
    2.20
     restores
    2.15
    restore
    2.10
     Restoration
    2.07
    Restore
    1.95
    Restoration
    1.94
    Activations Density 0.109%

    No Known Activations